INDEX

Explanations

words indicating corruption and moral judgments about individuals and groups

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.06

1:0.02

2:0.08

3:0.35

4:0.02

5:0.11

6:0.02

7:0.08

8:0.02

9:0.01

10:0.15

11:0.02

Negative Logits

 unfocusedRange

-1.97

title

-1.82

profile

-1.82

west

-1.80

otiation

-1.79

rower

-1.74

ridor

-1.70

Switch

-1.69

 Debate

-1.65

esthesia

-1.64

POSITIVE LOGITS

 robbers

3.24

 sinners

3.24

 parasites

3.17

 murderers

3.10

 idiots

3.08

 criminals

3.05

 villains

3.03

 invaders

2.97

 beings

2.93

 killers

2.86

Activations Density 0.631%

words indicating corruption and moral judgments about individuals and groups

No Comments

No Known Activations